good-first-issue-author
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- Command Execution: The skill executes local Python modules (
setup_preflight) and the GitHub CLI (gh). These commands are used to verify the skill's environment and to file the drafted issues upon maintainer confirmation.\n- Indirect Prompt Injection: The skill ingests untrusted data from issue descriptions and external backlog items. While this presents an entry point for untrusted content, the skill incorporates several mitigating factors.\n - Ingestion points: Processes candidate tasks, gap descriptions, and linked issues as described in
SKILL.md.\n - Boundary markers: The instructions explicitly direct the agent to treat external content as data rather than instructions and to flag potential injection attempts.\n
- Capability inventory: The skill possesses the capability to read repository files and create new issues using the
ghtool.\n - Sanitization: It employs a suitability gate to screen for security-sensitive topics and a nine-point readiness checklist to validate the drafted content before it is shown to the user.\n- External Downloads: The skill refers to a setup process (
/magpie-setup config) that installs project-specific checkers. This represents a standard initialization step for the framework to ensure all required configurations are present.
Audit Metadata